Abstract

Aims

The preliminary evidence supports an association between obstructive sleep apnoea (OSA), disturbed cardiac repolarization, and consequent cardiac dysrhythmias. The aim of the current trial was to assess the effects of continuous positive airway pressure (CPAP) therapy withdrawal on the measures of cardiac repolarization in patients with OSA.

Methods and results

Forty-one OSA patients established on CPAP treatment were randomized to either CPAP withdrawal (subtherapeutic CPAP) or continue therapeutic CPAP for 2 weeks. Polysomnography was performed, and indices of cardiac repolarization (QTc, TpTec intervals) and dispersion of repolarization (TpTe/QT ratio) were derived from 12-lead electrocardiography (ECG) at baseline and 2 weeks. Continuous positive airway pressure withdrawal led to a recurrence of OSA. Compared with therapeutic CPAP, subtherapeutic CPAP for 2 weeks was associated with a significant increase in the length of the QTc and TpTec intervals (mean difference between groups 21.4 ms, 95% CI 11.3–1.6 ms, P < 0.001 and 14.4 ms, 95% CI 7.2–21.5 ms, P < 0.001, respectively) and in the TpTe/QT ratio (mean difference between groups 0.02, 95% CI 0.00–0.03, P = 0.020). There was a statistically significant correlation between the change in apnoea/hypopnoea index (AHI) from baseline, and both the change in the QTc interval and the TpTec interval (r = 0.60, 95% CI 0.36–0.77, P < 0.001 and r = 0.45, 95% CI 0.17–0.67, P = 0.003, n = 41, respectively).

Conclusion

Continuous positive airway pressure withdrawal is associated with the prolongation of the QTc and TpTec intervals and TpTe/QT ratio, which may provide a possible mechanistic link between OSA, cardiac dysrhythmias, and thus sudden cardiac death.

Introduction

Obstructive sleep apnoea (OSA) is a highly prevalent sleep-related breathing disorder affecting up to 30% of middle-aged adults.1,2 Obstructive sleep apnoea is characterized by the repetitive interruption of ventilation during sleep caused by the collapse of the upper airway, resulting in apnoeas and hypopnoeas associated with oxygen desaturations and arousals from sleep. As a consequence of the fragmented sleep, patients with OSA may suffer from increased daytime sleepiness leading to impaired quality of life and the increased risk of traffic accidents.3,4

Symptomatic OSA has been implicated as an important factor in the pathogenesis of hypertension and cardiovascular disease.4–6 Furthermore, the findings of observational studies suggest that OSA may be causally related with cardiac dysrhythmias, such as premature ventricular contractions, atrial fibrillation (AF), ventricular arrhythmias, atrioventricular block, and sinus arrest.7–12 The observation that patients with OSA have an increased risk of sudden cardiac death (SCD) at night suggests an association between OSA and SCD.12 The mechanistic link between OSA and SCD, however, has not yet been thoroughly established. An increase in the dispersion of cardiac repolarization has been suggested as a potential mechanism of ventricular arrhythmias and SCD.13–15 The various markers derived from the surface electrocardiography (ECG) as well as from the electrophysiological studies provide information about the transmural dispersion of repolarization and the electrical inhomogeneity in the ventricles during repolarization.16–18

There is preliminary evidence from uncontrolled studies suggesting that cardiac repolarization is prolonged and the transmural dispersion of repolarization is increased, in patients with OSA. These ECG abnormalities seem to reverse with continuous positive airway pressure (CPAP) treatment.19 However, there is no data from randomized controlled interventional trials proving this association, and thus a controversy remains as to whether OSA is an independent aetiologic risk factor for disturbed cardiac repolarization.11

To address this uncertainty, we have analysed the ECG data from a randomized controlled trial exploring the effects of a 2-week CPAP withdrawal vs. continuing CPAP in patients with previously treated OSA,20 in order to investigate the effects of CPAP withdrawal and a return of OSA on ECG-derived measures of cardiac repolarization.

Methods

Patients, trial design, and sample calculation

Patients previously diagnosed with OSA, and treated with CPAP, who were registered in a database of the Sleep Disorders Centre and Pulmonary Division of the University Hospital of Zurich, Switzerland, were eligible for the trial if they were aged between 20 and 75 years, had an oxygen desaturation index (ODI, ≥4% dips) of >10/h during their initial sleep study, an ODI of >10/h during an ambulatory nocturnal pulse-oximetry performed on the last night of a four-night period without CPAP prior to the main study, and if they had been treated with CPAP for >12 months with a average compliance of ≥4 h per night.

Patients with previous ventilatory failure, Cheyne–Stokes breathing, unstable and untreated coronary or peripheral artery disease, severe and inadequately controlled arterial hypertension, a history of any sleep-related accident, or with a heavy goods vehicle driver's licence, were excluded from the study. The trial was performed according to the Declaration of Helsinki, was approved by the University Hospital of Zurich research ethics committee (EK-1600) and registered (ISRCTN 93153804). Written informed consent was obtained from all participants. The primary outcomes of this trial [change in the ODI and apnoea/hypopnoea index (AHI)] are reported elsewhere in more detail.20

After confirming the persistence of OSA by home overnight pulse-oximetry at the end of a four-night period without CPAP, eligible patients returned to therapy with CPAP for ≥7 days. After baseline assessments, the patients were randomized to either continue with CPAP therapy or switch to subtherapeutic CPAP for 2 weeks by a series of presealed and numbered envelopes. The follow-up assessments of ECG were performed at 1 and 2 weeks. Patients remained blinded as to whether they were receiving therapeutic or subtherapeutic CPAP, as did the investigators. The member of the research team who randomized the patients did not take part in the outcome assessments.

A sample size estimation was performed based on the assumption that a clinically relevant difference in the ODI (≥4% drop of oxygen saturation/h) between active treatment and placebo is 10/h (SD 10) and a clinically relevant difference in the QTc is 20 ms (SD 15).21 Based on these assumptions, the power calculation indicated that 41 patients were required in total in order to not miss clinically relevant differences in the ODI (with a power of 90%) and in the QTc (with a power of 99%).

Sleep studies and continuous positive airway pressure

At baseline and 2 weeks, attended polysomnographic sleep studies were performed and analysed according to the standard methods. Apnoeas were defined as a reduction in the amplitude of chest wall motion by >90% from baseline over the previous 2 min for >10 s, hypopnoeas were defined as a reduction in the amplitude of chest wall motion by >50% from baseline over the previous 2 min for >10 s, associated with a ≥4% drop in oxygen saturation.22 The severity of sleep-disordered breathing was quantified as the number of apnoeas/hypopnoeas (AHI) and oxygen desaturations ≥4% per hour of study (ODI). In patients randomized to subtherapeutic CPAP, the subtherapeutic pressure was achieved by setting the CPAP machine to the lowest pressure, inserting a flow-restricting connector at the machine outlet, and inserting six extra holes in the collar of the main tubing at the end of the mask to allow air escape, thus preventing rebreathing of carbon dioxide, as previously described.23 Sleepiness was assessed using the Epworth sleepiness score (ESS).24,25

Electrocardiography

The participants were asked to abstain from alcohol, tobacco, or caffeine on the day the measurements were done. The room temperature and lighting were set at the same level for all measurements. The participants rested for 5 min in the supine position before measurements were done. For all electrocardiographic recordings a commercially available 12-lead ECG (AT 104 PC, Schiller-Reomed AG, Switzerland) was used and set at 25-mm/s paper speed and 10-mm/mV amplitude. The measurements of the ECG intervals were performed in duplicate with a dedicated ECG analysis software (DatInf® Measure 2.1d, DatInf GmbH, Tübingen, Germany) by two investigators who were blinded to the patient's group allocation. The mean of the measurements by the two investigators was used for statistical analysis. Lead V5 was used and when unsuitable, lead V4 or V6. The indices of cardiac repolarization were determined as follows: the QT interval was defined as the time from the earliest onset of the QRS complex to the point at which the downward slope of the T-wave returned to baseline and the Tpeak-to-Tend interval (TpTe) as the time from the peak of the T-wave to the end of the T-wave.26 The end of the T-wave was defined as the cutting point of the tangent to the downward slope of the T-wave and the isoelectric line.18 Both the QT and TpTe intervals were corrected for the heart rate as described previously.27 As a measure of dispersion of repolarization, TpTe/QT ratio was calculated.14

Data analysis

All values are presented as mean (SD) unless otherwise stated. Statistical analysis was performed with Statistica V6.0 (StatSoft, Tulsa, OK, USA). Differences in baseline characteristics between the groups were assessed by independent t-tests and χ2 tests as appropriate. Comparisons of changes between groups were assessed by independent t-tests and Mann–Whitney U tests as appropriate. The analyses were performed on an intention-to-treat basis with no change assumed for the missing follow-up data. Correlations between changes in the indices of OSA severity and the ECG measures of cardiac repolarization were assessed by Pearson correlation coefficients. A two-sided P value of <0.05 was considered to be statistically significant.

Results

Trial profile and patient characteristics

Figure 1 shows the trial profile. The two groups were similar regarding all patients’ characteristics at baseline (Table 1). Data on AHI downloaded from the patients’ CPAP machines, as well as ESS values within the normal range, proved successful treatment of patients with CPAP prior to the study (Table 1). One patient in the subtherapeutic CPAP group withdrew from the study 4 days after randomization because of intolerable daytime symptoms.

Table 1

Baseline characteristics

 Therapeutic CPAP (n = 20) Subtherapeutic CPAP (n = 21) 
Age (years) 63.6 (5.1) 61.8 (7.5) 
Females (n1/20 0/21 
Body mass index (kg/m232.9 (6.5) 33.1 (4.4) 
Waist/hip circumference ratio 1.0 (0.1) 1.0 (0.0) 
Neck circumference (cm) 46.4 (3.7) 46.1 (4.2) 
Current smokers (n1/20 4/21 
Ex-smokers (n5/20 8/21 
Hypertension (n14/20 17/21 
Diabetes (n4/20 5/21 
CAD (n2/20 1/21 
Sokolow–Lyon index (mV) 21.1 (5.3) 21.1 (5.7) 
Cardiac dysrhythmiaa (n0/20 1/21 
Antihypertensive medication (n13/20 16/21 
 Beta-blockers (n7/20 6/21 
 Diuretics (n5/20 8/21 
Cholesterol-lowering medication (n7/20 7/21 
Glucose-lowering medication (n3/20 2/21 
Potassium (mmol/L) 3.9 (0.4) 3.9 (0.3) 
AHI original sleep study 36.0 (17.3) 45.3 (22.3) 
ODI original sleep study 26.6 (13.5) 37.3 (22.7) 
ODI 4-day withdrawal 25.4 (8.6) 28.9 (16.2) 
AHI on CPAPb 5.1 (2.7) 4.3 (2.3) 
CPAP compliance (min) 373.1 (67.9) 362.8 (72.3) 
CPAP use relative to time in bed (%)c 80.8 (14.5) 79.2 (20.0) 
Duration of CPAP treatment (years) 5.6 (3.4) 7.4 (3.2) 
ESS before therapy 7.4 (3.1) 6.6 (2.7) 
Time from screening to randomization (days) 75.9 (29.0) 68.3 (29.9) 
 Therapeutic CPAP (n = 20) Subtherapeutic CPAP (n = 21) 
Age (years) 63.6 (5.1) 61.8 (7.5) 
Females (n1/20 0/21 
Body mass index (kg/m232.9 (6.5) 33.1 (4.4) 
Waist/hip circumference ratio 1.0 (0.1) 1.0 (0.0) 
Neck circumference (cm) 46.4 (3.7) 46.1 (4.2) 
Current smokers (n1/20 4/21 
Ex-smokers (n5/20 8/21 
Hypertension (n14/20 17/21 
Diabetes (n4/20 5/21 
CAD (n2/20 1/21 
Sokolow–Lyon index (mV) 21.1 (5.3) 21.1 (5.7) 
Cardiac dysrhythmiaa (n0/20 1/21 
Antihypertensive medication (n13/20 16/21 
 Beta-blockers (n7/20 6/21 
 Diuretics (n5/20 8/21 
Cholesterol-lowering medication (n7/20 7/21 
Glucose-lowering medication (n3/20 2/21 
Potassium (mmol/L) 3.9 (0.4) 3.9 (0.3) 
AHI original sleep study 36.0 (17.3) 45.3 (22.3) 
ODI original sleep study 26.6 (13.5) 37.3 (22.7) 
ODI 4-day withdrawal 25.4 (8.6) 28.9 (16.2) 
AHI on CPAPb 5.1 (2.7) 4.3 (2.3) 
CPAP compliance (min) 373.1 (67.9) 362.8 (72.3) 
CPAP use relative to time in bed (%)c 80.8 (14.5) 79.2 (20.0) 
Duration of CPAP treatment (years) 5.6 (3.4) 7.4 (3.2) 
ESS before therapy 7.4 (3.1) 6.6 (2.7) 
Time from screening to randomization (days) 75.9 (29.0) 68.3 (29.9) 

Values are mean (SD) where applicable.

CAD, coronary heart disease; AHI, apnoea/hypopnoea index; ODI, oxygen desaturation index; AHI on CPAP, AHI downloaded from CPAP machine averaged from previous 6 months; ESS, Epworth sleepiness score.

aAtrial fibrillation.

bDerived from CPAP machine.

cTime in bed estimated by actimetry.

Figure 1

Trial profile.

Figure 1

Trial profile.

Sleep-disordered breathing and sleepiness

Compared with therapeutic CPAP, AHI increased significantly with the subtherapeutic CPAP at 2 weeks (mean difference in the AHI change 30.2, 95% CI 18.8–41.7) and ODI (mean difference in the ODI change 24.0, 95% CI 13.5–34.4, P < 0.001 for both comparisons). ESS increased significantly at 2 weeks in the subtherapeutic group compared with the therapeutic group (mean difference in ESS change 3.1, 95% CI 1.4–4.7, P < 0.001, Table 2).

Table 2

Effects of continuous positive airway pressure withdrawal on sleep-disordered breathing and sleepiness

  Baseline
 
Change from baseline
 
95% CI between groups P-value 
Therapeutic CPAP (n = 20) Subtherapeutic CPAP (n = 21) Therapeutic CPAP (n = 20) Subtherapeutic CPAP (n = 21) 
AHI (1/h) 1.7 (1.8) 2.2 (2.5) 0.4 (2.8) 30.6 (25.2) 18.8/41.7 <0.001 
ODI (1/h) 0.5 (0.8) 0.9 (2.0) −0.2 (1.0) 23.8 (23.1) 13.5/34.4 <0.001 
ESS 7.4 (3.1) 6.6 (2.7) −0.9 (2.2) 2.2 (2.9) 1.4/4.7 <0.001 
  Baseline
 
Change from baseline
 
95% CI between groups P-value 
Therapeutic CPAP (n = 20) Subtherapeutic CPAP (n = 21) Therapeutic CPAP (n = 20) Subtherapeutic CPAP (n = 21) 
AHI (1/h) 1.7 (1.8) 2.2 (2.5) 0.4 (2.8) 30.6 (25.2) 18.8/41.7 <0.001 
ODI (1/h) 0.5 (0.8) 0.9 (2.0) −0.2 (1.0) 23.8 (23.1) 13.5/34.4 <0.001 
ESS 7.4 (3.1) 6.6 (2.7) −0.9 (2.2) 2.2 (2.9) 1.4/4.7 <0.001 

Values are mean (SD). 95% CI = 95% confidence interval of the difference between groups in change from baseline.

AHI, apnoea/hypopnoea index; ODI, oxygen desaturation index; ESS, Epworth sleepiness score.

Effect of continuous positive airway pressure withdrawal on cardiac repolarization indices

ECG interval times at baseline, and changes from baseline, are shown in Table 3. There were no relevant differences between the two groups at baseline.

Table 3

Effect of continuous positive airway pressure withdrawal on the electrocardiographic intervals and urinary norepinephrine

  Baseline
 
Change from baseline
 
95% CI between groups P-value 
Therapeutic CPAP (n = 20) Subtherapeutic CPAP (n = 21) Therapeutic CPAP (n = 20) Subtherapeutic CPAP (n = 21) 
RR (ms) 962.2 (180.2) 892.8 (205.4) −46.9 (111.5) −54.7 (195.7) −92.2/108 0.875 
P (ms) 108.5 (15.3) 110.6 (16.2) −3.3 (34.7) 8.1 (46.2) −14.5/37.3 0.379 
PQ (ms) 196.7 (25.2) 181.1 (26.3) −1.8 (18.4) −2.9 (11.7) −10.8/8.6 0.824 
QRS (ms) 106.4 (25.6) 101.0 (13.7) 0.9 (15.9) 1.1 (11.2) −8.4/8.9 0.960 
QT (ms) 416.8 (38.5) 389.1 (39.2) −11.5 (38.6) 9.0 (18.7) 1.5/39.5 0.035 
QTc (ms)a 421.9 (16.0) 416.1 (28.0) −0.1 (9.3) 21.3 (20.4) 11.3/31.6 <0.001 
TpTe (ms) 108.8 (14.4) 104.3 (12.6) −2.4 (10.7) 9.6 (10.9) 7.0/16.9 0.001 
TpTec (ms)a 111.8 (14.5) 114.7 (14.6) 0.5 (11.3) 14.9 (11.3) 7.2/21.5 <0.001 
TpTe/QT 0.26 (0.03) 0.27 (0.04) 0.00 (0.03) 0.02 (0.03) −0.00/0.03 0.020 
Norepinephrine (nmoL/mmoL) 22.1 (11.5) 26.3 (10.2) 0.9 (5.4) 11.5 (15.5) 2.5/18.7 0.012 
  Baseline
 
Change from baseline
 
95% CI between groups P-value 
Therapeutic CPAP (n = 20) Subtherapeutic CPAP (n = 21) Therapeutic CPAP (n = 20) Subtherapeutic CPAP (n = 21) 
RR (ms) 962.2 (180.2) 892.8 (205.4) −46.9 (111.5) −54.7 (195.7) −92.2/108 0.875 
P (ms) 108.5 (15.3) 110.6 (16.2) −3.3 (34.7) 8.1 (46.2) −14.5/37.3 0.379 
PQ (ms) 196.7 (25.2) 181.1 (26.3) −1.8 (18.4) −2.9 (11.7) −10.8/8.6 0.824 
QRS (ms) 106.4 (25.6) 101.0 (13.7) 0.9 (15.9) 1.1 (11.2) −8.4/8.9 0.960 
QT (ms) 416.8 (38.5) 389.1 (39.2) −11.5 (38.6) 9.0 (18.7) 1.5/39.5 0.035 
QTc (ms)a 421.9 (16.0) 416.1 (28.0) −0.1 (9.3) 21.3 (20.4) 11.3/31.6 <0.001 
TpTe (ms) 108.8 (14.4) 104.3 (12.6) −2.4 (10.7) 9.6 (10.9) 7.0/16.9 0.001 
TpTec (ms)a 111.8 (14.5) 114.7 (14.6) 0.5 (11.3) 14.9 (11.3) 7.2/21.5 <0.001 
TpTe/QT 0.26 (0.03) 0.27 (0.04) 0.00 (0.03) 0.02 (0.03) −0.00/0.03 0.020 
Norepinephrine (nmoL/mmoL) 22.1 (11.5) 26.3 (10.2) 0.9 (5.4) 11.5 (15.5) 2.5/18.7 0.012 

Values are mean (SD).

95% CI = 95% confidence interval of the difference between groups in change from baseline. N = 17 and 18 for urinary norepinephrine (nmoL/mmoL of creatinine) in the therapeutic and subtherapeutic CPAP group, respectively.

aHeart-rate corrected.

Compared with therapeutic CPAP, subtherapeutic CPAP for 2 weeks was associated with a significant increase in the length of the QTc and TpTec intervals (mean difference 21.4 ms, 95% CI 11.3–31.6 ms, P < 0.001 and 14.4 ms, 95% CI 7.2–21.5 ms, P < 0.001, respectively) (Figures 2 and 3) and in the TpTe/QT ratio (mean difference 0.02, 95% CI −0.00–0.03, P = 0.020).

Figure 2

Data are presented as mean and standard error. Black bars represent the subtherapeutic continuous positive airway pressure group and white bars represent the therapeutic continuous positive airway pressure group. There was a significant increase in the length of the QTc at 2 weeks in the subtherapeutic continuous positive airway pressure group compared with the therapeutic continuous positive airway pressure group. *P < 0.001 for comparisons of changes from baseline between groups.

Figure 2

Data are presented as mean and standard error. Black bars represent the subtherapeutic continuous positive airway pressure group and white bars represent the therapeutic continuous positive airway pressure group. There was a significant increase in the length of the QTc at 2 weeks in the subtherapeutic continuous positive airway pressure group compared with the therapeutic continuous positive airway pressure group. *P < 0.001 for comparisons of changes from baseline between groups.

Figure 3

Data are presented as mean and standard error. Black bars represent the subtherapeutic continuous positive airway pressure group and white bars represent the therapeutic continuous positive airway pressure group. There was a significant increase in the length of the TpTec at 2 weeks in the subtherapeutic continuous positive airway pressure group compared with the therapeutic continuous positive airway pressure group. *P < 0.001 for comparisons of changes from baseline between groups.

Figure 3

Data are presented as mean and standard error. Black bars represent the subtherapeutic continuous positive airway pressure group and white bars represent the therapeutic continuous positive airway pressure group. There was a significant increase in the length of the TpTec at 2 weeks in the subtherapeutic continuous positive airway pressure group compared with the therapeutic continuous positive airway pressure group. *P < 0.001 for comparisons of changes from baseline between groups.

A statistically significant correlation was observed between the change in the AHI, and both the changes in the QTc and the TpTec intervals (r = 0.60, 95% CI 0.36–0.77, P < 0.001 and r = 0.45, 95% CI 0.17–0.67, P = 0.003, respectively, n = 41, Figures 4 and 5).

Figure 4

Correlation between the change in the QTc interval (change from baseline to 2 weeks) and change in the apnoea/hypopnoea index (r = 0.60, 95% CI 0.36–.77, P < 0.001, n = 41). Black circles represent patients in the subtherapeutic continuous positive airway pressure group and white circles represent patients in the therapeutic continuous positive airway pressure group.

Figure 4

Correlation between the change in the QTc interval (change from baseline to 2 weeks) and change in the apnoea/hypopnoea index (r = 0.60, 95% CI 0.36–.77, P < 0.001, n = 41). Black circles represent patients in the subtherapeutic continuous positive airway pressure group and white circles represent patients in the therapeutic continuous positive airway pressure group.

Figure 5

Correlation between the change in the TpTec interval (change from baseline to 2 weeks) and change in the apnoea/hypopnoea index (r = 0.45, 95% CI 0.17–0.67, P = 0.003, n = 41). Black circles represent patients in the subtherapeutic continuous positive airway pressure group and white circles represent patients in the therapeutic continuous positive airway pressure group.

Figure 5

Correlation between the change in the TpTec interval (change from baseline to 2 weeks) and change in the apnoea/hypopnoea index (r = 0.45, 95% CI 0.17–0.67, P = 0.003, n = 41). Black circles represent patients in the subtherapeutic continuous positive airway pressure group and white circles represent patients in the therapeutic continuous positive airway pressure group.

A statistically significant correlation was also found between the change in the ODI, and both the changes in the QTc and the TpTec intervals (r = 0.44, 95% CI 0.16–0.66, P = 0.004, and r = 0.51, 95% CI 0.24–0.71, P < 0.001, respectively, n = 41).

Discussion

This is the first randomized controlled trial providing data on the effects of OSA on ECG measures of cardiac repolarization. We found that CPAP withdrawal was associated with a considerable prolongation and an increase in the dispersion of cardiac repolarization. These findings provide further evidence suggesting a possible link between OSA, cardiac dysrhythmias, and SCD.

The QT interval represents the electrocardiographic correlate of ventricular de- and repolarization, including the vulnerable period for reentry tachycardia. As such, numerous studies have identified the inherited as well as acquired QT prolongation as a risk factor for the occurrence of malignant cardiac dysrhythmias and SCD.28,29

The Tpeak-to-Tend (TpTe) interval is a measure of cardiac transmural dispersion of repolarization, which is explained by a gradient of action potential duration from endo- (longest) to epicardial cells (shortest).30,31 A prolongation of the TpTe interval leads to increased vulnerability for the occurrence of early after-depolarizations and has been associated with ventricular tachycardia and an increased risk for SCD.32–34 Specifically, recent findings from Panikkath et al.33 suggest that patients with an uncorrected TpTe interval >100 ms in the resting ECG are at increased risk of SCD.

In the current study, QTc and TpTec intervals at baseline were already considerably longer in the CPAP-treated study population than in a group of healthy young adults in one of our previous studies35 (416.1 vs. 403.8 and 114.7 vs. 87.3 ms, respectively) possibly due to a high rate of comorbidities seen in our OSA population. However, we found that CPAP withdrawal for 2 weeks was associated with further, statistically significant prolongation of the QTc and TpTec intervals. The withdrawal of CPAP was also associated with an increase in the uncorrected mean TpTe interval from 104 to 114 ms; as a result, 71% of patients in the subtherapeutic CPAP group were shifted to TpTe values >100 ms and 55% were shifted to QTc values >430 ms. Since these cut-offs were previously associated with an elevated risk for SCD,33 our data imply that many OSA patients without CPAP treatment may fall into this category. Moreover, our data demonstrate an increase in the effect of CPAP withdrawal on cardiac repolarization parameters across 2 weeks (Figures 2 and 3), indicating that further withdrawal may result in an even more pronounced increase in these values.

The increase in the length of the QTc and TpTec intervals was positively correlated with the change in the severity of sleep-disordered breathing (Figures 4 and 5) suggesting that the risk for SCD increases with the severity of OSA. There is also preliminary data from uncontrolled studies in patients with OSA showing that QT intervals are longer during the night-time compared with the daytime,21 and longer at the onset of an apnoea episode compared with the period after an apnoea.36 In an observational study, OSA has been shown to be associated with the increased QT interval variability during sleep, possibly reflecting alterations in the sympathetic outflow to the ventricular myocardium.37

In addition to the prolonged QTc and TpTec intervals, an increased TpTe/QT ratio, a measure of disproportional prolongation of global dispersion relative to the QT interval, may have an important role in arrhythmogenesis.16 This ratio has the advantage of more reliably eliminating the confounding effects of the heart rate variability in the ECG and the inter-individual variation in the length of the QT interval.14 We were able to demonstrate an increase in the TpTe/QT ratio after CPAP withdrawal, further substantiating a potential risk for arrhythmias in this situation. These findings are in line with the preliminary evidence from observational studies, suggesting that the severity of OSA is also associated with the worsening of QT dynamicity parameters (QT/RR slope), an indicator of abnormal rate adaption of ventricular repolarization.19,37 This may be due to a relative lack of parasympathetic tone in patients with OSA and, thus, be a predictor of myocardial vulnerability and tendency to cardiac dysrhythmias.19,38

A prolongation of QRS duration, i.e. a prolongation of ventricular depolarization, has been suggested to be associated with an increased propensity for SCD.39 In our study, we found a small increase in QRS duration in the CPAP withdrawal group, which was, however, not statistically significant (1.1 ms, P = 0.96). The risk of SCD associated with QRS prolongation is mainly interpreted as a result of overt coronary artery disease (ischaemia or scar).40,41 However, in our study population, only three patients (two in the CPAP group and one in the CPAP withdrawal group) had (stable) coronary artery disease. Thus the chances of finding a significant prolongation of the QRS complex during the CPAP withdrawal were likely to be much smaller than it would have been in patients with overt coronary heart disease. This is in contrast to the more subtle markers of repolarization such as TpTec. It hence remains unclear at present whether OSA directly affects ventricular depolarization in the longer term, and further studies are required to conclusively answer this question.

There may be several mechanisms explaining the disturbance of repolarization and the high incidence of atrial and ventricular dysrhythmias and SCD in OSA patients. Obstructive apnoea and hypopnoea are associated with repeated inspiratory efforts against the collapsed upper airways associated with large swings in the intrathoracic pressure, which may be as high as 60–80 mmHg. During each obstructive apnoea and hypopnoea, the negative intrathoracic pressure exerts a direct distending force on the heart and aorta.42 Cardiac wall stress is additionally augmented by considerable acute changes in the atrial and ventricular volume during obstructive apnoea and hypopnoea.43–45In vitro experiments on myocardial tissue have shown that acute atrial stretch (similar to the atrial stretch occurring during apnoea/hypopnoea) induces both early and delayed after-depolarizations which may trigger premature myocardial contractions.44,45 In an animal model, Linz et al.46 reported shortening of the right atrial refractory period and the increased susceptibility to premature beats and AF induced by negative thoracic pressure during obstructive respiratory events. This mechanism, repeated during each apnoeic phase, may lead in the long-term to the electrical and mechanical remodelling of both the atria and left ventricle,47,48 thereby possibly explaining the high prevalence of AF in patients with OSA.49,50

A further possible mechanism for the alteration of cardiac repolarization may be autonomic nervous system imbalance which has been shown to be present in OSA patients.6 Baumert et al.51 showed that experimentally induced increased cardiac sympathetic activity is positively correlated with QT variability in hypertensive patients. We previously reported an increased sympathetic activity as assessed by the urinary noradrenaline (mean increase 43.7%) and heart rate (mean increase 9.7%) after 2 weeks of CPAP withdrawal in the current study population.20 In addition, there was a correlation between the change in the QTc and change in the heart rate (r = 0.45, P = 0.003, data not shown) as well as between the change in the QTc and change in the urinary noradrenaline levels (r = 0.37, P = 0.020, data not shown). These findings suggest that increased sympathetic nervous system activity may be one of the underlying mechanisms between OSA and the increased dispersion of cardiac repolarization, but further studies are warranted.

In summary, the increased susceptibility to dysrhythmias may therefore be explained by structural remodelling (resulting in the formation of an arrhythmogenic substrate),48 electrical remodelling (manifesting as disturbed repolarization), increased sympathetic activation, and triggers (including acute episodes of apnoea and hypopnoea), which may eventually lead to the onset of ventricular dysrhythmias (Coumel's triangle).47

The current study has some inherent limitations. We performed standard resting 12-lead ECG and not continuous 24-h ECG, thus it is not possible to determine from the current study, the effects of CPAP withdrawal on ECG for longer periods, or during sleep. Therefore, further data from randomized controlled studies are needed to evaluate the effect of CPAP on measures of repolarization during sleep and longer daytime periods. Future interventional studies will also need to address the question as to which mechanisms underpin the association between OSA and both the prolongation and the increased dispersion of cardiac repolarization. The question as to whether OSA is associated with chronic and irreversible alterations of the ECG will need to be addressed in future trials with conventional design, however, due to ethical issues related to a long-term control group, it will be very difficult to perform a randomized controlled trial to answer this question. In addition, the findings of a study that selects patients with high CPAP compliance cannot be generalized without caution to the untreated OSA population usually seen in a sleep clinic.

In conclusion, we have found that CPAP withdrawal for 2 weeks was associated with prolongation as well as increased dispersion of cardiac repolarization. These findings may provide a possible mechanistic link between OSA, cardiac arrhythmias, and SCD.

Funding

This work was supported by the Swiss National Science Foundation (32003B_124915), the Swiss Society of Pneumology, and the University of Zurich (Matching Funds).

Conflict of interest: None declared.

References

1
Stradling
JR
Crosby
JH
Predictors and prevalence of obstructive sleep apnoea and snoring in 1001 middle aged men
Thorax
 , 
1991
, vol. 
46
 (pg. 
85
-
90
)
2
Young
T
Palta
M
Dempsey
J
Skatrud
J
Weber
S
Badr
S
The occurrence of sleep-disordered breathing among middle-aged adults
N Engl J Med
 , 
1993
, vol. 
328
 (pg. 
1230
-
1235
)
3
George
CF
Reduction in motor vehicle collisions following treatment of sleep apnoea with nasal CPAP
Thorax
 , 
2001
, vol. 
56
 (pg. 
508
-
512
)
4
Kohler
M
Stradling
JR
Mechanisms of vascular damage in obstructive sleep apnea
Nat Rev Cardiol
 , 
2010
, vol. 
7
 (pg. 
677
-
685
)
5
Marin
JM
Carrizo
SJ
Vicente
E
Agusti
AG
Long-term cardiovascular outcomes in men with obstructive sleep apnoea–hypopnoea with or without treatment with continuous positive airway pressure: an observational study
Lancet
 , 
2005
, vol. 
365
 (pg. 
1046
-
1053
)
6
Kohler
M
Pepperell
JC
Casadei
B
Craig
S
Crosthwaite
N
Stradling
JR
Davies
RJ
CPAP and measures of cardiovascular risk in males with OSAS
Eur Respir J
 , 
2008
, vol. 
32
 (pg. 
1488
-
1496
)
7
Guilleminault
C
Connolly
SJ
Winkle
RA
Cardiac arrhythmia and conduction disturbances during sleep in 400 patients with sleep apnea syndrome
Am J Cardiol
 , 
1983
, vol. 
52
 (pg. 
490
-
494
)
8
Namtvedt
SK
Randby
A
Einvik
G
Hrubos-Strøm
H
Somers
VK
Røsjø
H
Omland
T
Cardiac arrhythmias in obstructive sleep apnea (from the Akershus Sleep Apnea Project)
Am J Cardiol
 , 
2011
, vol. 
108
 (pg. 
1141
-
1146
)
9
Hoffstein
V
Mateika
S
Cardiac arrhythmias, snoring, and sleep apnea. 1994
. Chest
 , 
2009
, vol. 
136
 pg. 
e30
 
10
Gami
AS
Pressman
G
Caples
SM
Kanagala
R
Gard
JJ
Davison
DE
Malouf
JF
Ammash
NM
Friedman
PA
Somers
VK
Association of atrial fibrillation and obstructive sleep apnea
Circulation
 , 
2004
, vol. 
110
 (pg. 
364
-
367
)
11
Somers
VK
White
DP
Amin
R
Abraham
WT
Costa
F
Culebras
A
Daniels
S
Floras
JS
Hunt
CE
Olson
LJ
Pickering
TG
Russell
R
Woo
M
Young
T
Sleep apnea and cardiovascular disease
Circulation
 , 
2008
, vol. 
118
 (pg. 
1080
-
1111
)
12
Gami
AS
Howard
DE
Olson
EJ
Somers
VK
Day–night pattern of sudden death in obstructive sleep apnea
N Engl J Med
 , 
2005
, vol. 
352
 (pg. 
1206
-
1214
)
13
Kors
JA
Ritsema van Eck
HJ
van
HG
The meaning of the Tp–Te interval and its diagnostic value
J Electrocardiol
 , 
2008
, vol. 
41
 (pg. 
575
-
580
)
14
Gupta
P
Patel
C
Patel
H
Narayanaswamy
S
Malhotra
B
Green
JT
Yan
GX
T(p–e)/QT ratio as an index of arrhythmogenesis
J Electrocardiol
 , 
2008
, vol. 
41
 (pg. 
567
-
574
)
15
Castro Hevia
J
Antzelevitch
C
Tornés Bárzaga
F
Dorantes Sánchez
M
Dorticós Balea
F
Zayas Molina
R
Quinhones Pérez
MA
Fayad Rodríguez
Y
Tpeak–Tend and Tpeak–Tend dispersion as risk factors for ventricular tachycardia/ventricular fibrillation in patients with the Brugada syndrome
J Am Coll Cardiol
 , 
2006
, vol. 
47
 (pg. 
1828
-
1834
)
16
Yan
GX
Antzelevitch
C
Cellular basis for the normal T wave and the electrocardiographic manifestations of the long-QT syndrome
Circulation
 , 
1998
, vol. 
98
 (pg. 
1928
-
1936
)
17
Antzelevitch
C
Fish
J
Electrical heterogeneity within the ventricular wall
Basic Res Cardiol
 , 
2001
, vol. 
96
 (pg. 
517
-
527
)
18
Antzelevitch
C
T peak-Tend interval as an index of transmural dispersion of repolarization
Eur J Clin Invest
 , 
2001
, vol. 
31
 (pg. 
555
-
557
)
19
Roche
F
Barthelemy
JC
Garet
M
Duverney
D
Pichot
V
Sforza
E
Continuous positive airway pressure treatment improves the QT rate dependence adaptation of obstructive sleep apnea patients
Pacing Clin Electrophysiol
 , 
2005
, vol. 
28
 (pg. 
819
-
825
)
20
Kohler
M
Stoewhas
AC
Ayers
L
Senn
O
Bloch
KE
Russi
EW
Stradling
JR
The effects of CPAP therapy withdrawal in patients with obstructive sleep apnea: a randomised controlled trial
Am J Respir Crit Care Med
 , 
2011
, vol. 
184
 (pg. 
1192
-
1199
)
21
Barta
K
Szabó
Z
Kun
C
Munkácsy
C
Bene
O
Magyar
M
Csiba
L
Lörincz
I
The effect of sleep apnea on QT interval, QT dispersion and arrhythmias
Clin Cardiol
 , 
2009
, vol. 
33
 (pg. 
E35
-
E39
)
22
Peppard
PE
Young
T
Palta
M
Skatrud
J
Prospective study of the association between sleep-disordered breathing and hypertension
N Engl J Med
 , 
2000
, vol. 
342
 (pg. 
1378
-
1384
)
23
Pepperell
JC
Ramdassingh-Dow
S
Crosthwaite
N
Mullins
R
Jenkinson
C
Stradling
JR
Davies
RJ
Ambulatory blood pressure after therapeutic and subtherapeutic nasal continuous positive airway pressure for obstructive sleep apnoea: a randomised parallel trial
Lancet
 , 
2002
, vol. 
359
 (pg. 
204
-
210
)
24
Johns
MW
A new method for measuring daytime sleepiness: the Epworth sleepiness scale
Sleep
 , 
1991
, vol. 
14
 (pg. 
540
-
545
)
25
Bloch
KE
Schoch
OD
Zhang
JN
Russi
EW
German version of the Epworth Sleepiness Scale
Respiration
 , 
1999
, vol. 
66
 (pg. 
40
-
447
)
26
Lepeschkin
E
Surawicz
B
The measurement of the Q–T interval of the electrocardiogram
Circulation
 , 
1952
, vol. 
6
 (pg. 
378
-
388
)
27
Bazett
HC
The time relations of the blood-pressure changes after excision of the adrenal glands, with some observations on blood volume changes
J Physiol
 , 
1920
, vol. 
53
 (pg. 
320
-
339
)
28
Elming
H
Brendorp
B
Kober
L
Sahebzadah
N
Torp-Petersen
C
QTc interval in the assessment of cardiac risk
Card Electrophysiol Rev
 , 
2002
, vol. 
6
 (pg. 
289
-
294
)
29
Viskin
S
Long QT syndromes and torsade de pointes
Lancet
 , 
1999
, vol. 
354
 (pg. 
1625
-
1633
)
30
Antzelevitch
C
Cellular basis for the repolarization waves of the ECG
Ann N Y Acad Sci
 , 
2006
, vol. 
1080
 (pg. 
268
-
281
)
31
Xia
Y
Liang
Y
Kongstad
O
Holm
M
Olsson
B
Yuan
S
Tpeak–Tend interval as an index of global dispersion of ventricular repolarization: evaluations using monophasic action potential mapping of the epi- and endocardium in swine
J Interv Card Electrophysiol
 , 
2005
, vol. 
14
 (pg. 
79
-
87
)
32
Yamaguchi
M
Shimizu
M
Ino
H
Terai
H
Uchiyama
K
Oe
K
Mabuchi
T
Konno
T
Kaneda
T
Mabuchi
H
T wave peak-to-end interval and QT dispersion in acquired long QT syndrome: a new index for arrhythmogenicity
Clin Sci (Lond)
 , 
2003
, vol. 
105
 (pg. 
671
-
676
)
33
Panikkath
R
Reinier
K
Uy-Evanado
A
Teodorescu
C
Hattenhauer
J
Mariani
R
Gunson
K
Jui
J
Chugh
SS
Prolonged Tpeak-to-tend interval on the resting ECG is associated with increased risk of sudden cardiac death
Circ Arrhythm Electrophysiol
 , 
2011
, vol. 
4
 (pg. 
441
-
447
)
34
Topilski
I
Rogowski
O
Rosso
R
Justo
D
Copperman
Y
Glikson
M
Belhassen
B
Hochenberg
M
Viskin
S
The morphology of the QT interval predicts torsade de pointes during acquired bradyarrhythmias
J Am Coll Cardiol
 , 
2007
, vol. 
49
 (pg. 
320
-
328
)
35
Camen
G
Clarenbach
CF
Stöwhas
AC
Rossi
VA
Sievi
NA
Stradling
JR
Kohler
M
The effect of simulated obstructive apnea and hypopnea on heart rhythm
Swiss Arch Neurol Psych
 , 
2011
, vol. 
162
 
Suppl. 4
pg. 
19S
  
(abstract)
36
Gillis
AM
Stoohs
R
Guilleminault
C
Changes in the QT interval during obstructive sleep apnea
Sleep
 , 
1991
, vol. 
14
 (pg. 
346
-
350
)
37
Baumert
M
Schlaich
MP
Nalivaiko
E
Lambert
E
Sari
CI
Kaye
DM
Elser
MD
Sanders
P
Lambert
G
Relation between QT interval variability and cardiac sympathetic activity in hypertension
Am J Physiol Heart Circ Physiol
 , 
2011
, vol. 
300
 (pg. 
1412
-
1417
)
38
Roche
F
Gaspoz
JM
Court-Fortune
I
Costes
F
Geyssant
A
Duverney
D
Pichot
V
Barthélémy
JC
Alteration of QT rate dependence reflects cardiac autonomic imbalance in patients with obstructive sleep apnea syndrome
Pacing Clin Electrophysiol
 , 
2003
, vol. 
26
 (pg. 
1446
-
1453
)
39
Goldberger
JJ
Cain
ME
Hohnloser
SH
Kadish
AH
Knight
BP
Lauer
MS
Maron
BJ
Page
RL
Passman
RS
Siscovick
D
Stevenson
WG
Zipes
DP
American Heart Association/American College of Cardiovasculary Foundation/Heart Rhythm Society Scientific statement on noninvasive risk stratification techniques for identifying patients at risk for sudden cardiac death
Circulation
 , 
2008
, vol. 
118
 (pg. 
1497
-
1518
)
40
Brenyo
A
Zareba
W
Prognostic significance of QRS duration and morphology
Cardiol J
 , 
2011
, vol. 
18
 (pg. 
8
-
17
)
41
Englbom
H
Wagner
GS
Setser
RM
Selvester
RH
Billgren
T
Kasper
JM
Maynard
C
Pahlm
O
Arheden
H
White
RD
Quantitative clinical assessment of chronic anterior myocardial infarction with delayed enhancement magnetic resonance imaging and QRS scoring
Am Heart J
 , 
2003
, vol. 
146
 (pg. 
359
-
366
)
42
Stoewhas
AC
Namdar
M
Biaggi
P
Russi
EW
Bloch
KE
Stradling
JR
Kohler
M
The effect of simulated obstructive apnea and hypopnea on aortic diameter and BP
Chest
 , 
2011
, vol. 
140
 (pg. 
675
-
680
)
43
Orban
M
Bruce
CJ
Pressman
GS
Leinveber
P
Romero-Corral
A
Korinek
J
Konecny
T
Villarraga
HR
Kara
T
Caples
SM
Somers
VK
Dynamic changes of left ventricular performance and left atrial volume induced by the mueller maneuver in healthy young adults and implications for obstructive sleep apnea, atrial fibrillation, and heart failure
Am J Cardiol
 , 
2008
, vol. 
102
 (pg. 
1557
-
1561
)
44
Virolainen
J
Ventila
M
Turto
H
Kupari
M
Effect of negative intrathoracic pressure on left ventricular pressure dynamics and relaxation
J Appl Physiol
 , 
1995
, vol. 
79
 (pg. 
455
-
460
)
45
Koshino
Y
Villarraga
HR
Orban
M
Bruce
CJ
Pressman
GS
Leinveber
P
Saleh
HK
Konecny
T
Kara
T
Somers
VK
Lopez-Jimenez
F
Changes in left and right ventricular mechanics during the Mueller maneuver in healthy adults: a possible mechanism for abnormal cardiac function in patients with obstructive sleep apnea
Circ Cardiovasc Imaging
 , 
2010
, vol. 
3
 (pg. 
282
-
289
)
46
Linz
D
Schotten
U
Neuberger
HR
Bohm
M
Wirth
K
Negative tracheal pressure during obstructive respiratory events promotes atrial fibrillation by vagal activation
Heart Rhythm
 , 
2011
, vol. 
8
 (pg. 
1436
-
1443
)
47
Chan
KH
Wilcox
I
Obstructive sleep apnea: novel trigger and potential therapeutic target for cardiac arrhythmias
Expert Rev Cardiovasc Ther
 , 
2010
, vol. 
8
 (pg. 
981
-
994
)
48
Dimitri
H
Ng
M
Books
AG
Kuklik
P
Stiles
MK
Lau
DH
Antic
N
Thornton
A
Saint
DA
Mc Evoy
D
Antic
R
Kalman
JM
Sanders
P
Atrial remodelling in obstructive sleep apnea: implications for atrial fibrillation
Heart Rhythm
 , 
2012
, vol. 
9
 (pg. 
321
-
327
)
49
Hoyer
FF
Lickfett
LM
Mittmann-Braun
E
Ruland
C
Kreuz
J
Pabst
S
Schrickel
J
Juergens
U
Nickenig
G
Skowasch
D
High prevalence of obstructive sleep apnea in patients with resistant paroxysmal atrial fibrillation after pulmonary vein isolation
J Interv Card Electrophysiol
 , 
2010
, vol. 
1
 (pg. 
37
-
41
)
50
Andreas
S
von Breska
B
Schaumann
A
Gonska
BD
Kreuzer
H
Obstructive sleep apnoea and signal averaged electrocardiogram
Eur Respir J
 , 
1995
, vol. 
8
 (pg. 
546
-
550
)
51
Baumert
M
Smith
J
Catcheside
P
McEvoy
DR
Abbott
D
Sander
P
Nalivaiko
E
Variability of QT interval duration in obstructive sleep apnoea: an indicator of disease severity
Sleep
 , 
2008
, vol. 
31
 (pg. 
959
-
966
)

Author notes

The work was conducted at Sleep Disorders Centre and Pulmonary Division, University Hospital of Zurich, Switzerland.
This paper was guest edited by Prof. Gerhard Hindricks, University Leipzig, Heart Center, Department of Electrophysiology, Strumpellstrasse 39, Leipzig D-04289, Germany.

Supplementary data

Comments

0 Comments